Skip to content

14. Testing your own code

Every lesson so far has had checks under its task. They are tests, and you can write your own. A test block has a name in quotes and holds assert lines. Each assert states something that should be True:

fn double(n: Int) -> Int
    n * 2
end

test "double"
    assert double(2) == 4
    assert double(5) == 10
end

When an assert is False, the test fails, and its report shows the value your code made. Say double had a mistake, n + 2 instead of n * 2. The same test would report:

FAIL main.nomi :: double
  line 7: assertion failed
    assert double(5) == 10
    values:
      double(5)
        = 7
test result: FAILED. 0 passed, 1 failed

It names the test and the line that failed, and shows that double(5) made 7, not 10.

Look at the first assert: double(2) == 4 is True even with the mistake, because 2 + 2 and 2 * 2 are both 4. A test that checks one value can pass by luck. Check a few, and include the edges, where the answer changes.

This time the function is right, and you write the tests. is_teen says whether someone is a teenager: 13 to 19, both included.

Check runs your tests twice: first on this is_teen, where they should all pass, and then on a copy with a mistake hidden in it, where at least one should fail. The test below passes on both, so it can’t tell them apart. Add asserts until yours do.

fn is_teen(age: Int) -> Bool
    age >= 13 and age <= 19
end

// Write your tests below. Check runs them on this is_teen and on one with a mistake.
test "a teen and a child"
    assert is_teen(15)
    assert not is_teen(10)
end